• by December 12, 2000

By Christopher Parente

Statistics trumpeting the growth of email marketing seem ubiquitous today. We are exhorted to add email to our marketing mix, and are told email as a percentage of online ad spending is exploding.

Many marketers are making email a larger part of their overall marketing efforts, and they should. We all know why email works on paper - pardon the analogy - it offers some potent advantages over other marketing channels, including a very fast response time, detailed tracking, and the possibility for quick implementation.

But from a customer acquisition standpoint, what kind of results can you really expect from an email campaign using third party lists? Recently I conducted two email campaigns for DSL.net, a national provider of high-speed Internet access (as you may have guessed from the name,) so let me share with you the actual process and the results we achieved.

Setting the Requirements

DSL.net focuses exclusively on business customers, often small- to medium-sized companies in second or third tier markets. Digital subscriber line technology is distance sensitive, with service limited to addresses roughly three miles or less from the nearest local telephone central office. So immediately there were challenges in finding the right list.

In email marketing today, there is a wider selection of lists in B-to-C than there is in B-to-B. Plus, the distance limitations of DSL dictated that unless additional screening was performed on a potential list, DSL.net would be sending emails to thousands of unserviceable addresses.

In traditional direct mail, the company deals with providers like Dun & Bradstreet and Marigold, and runs addresses through a proprietary software program that segments out the serviceable addresses from those outside the three-mile radius. So DSL.net needed the same functionality from an email partner.

The Process

This functionality was not easy to find. First off, some online marketing shops immediately want a new client to sign a one to three month contract, on a traditional retainer basis, rather than address the client's specific need. They want to get married before the first date! Other vendors could not accommodate the need to filter the addresses by location, either because the list owner would not release the lists, or the email address had no corresponding physical snail address. In some cases the vendor could do it, but the process eliminated so many addresses that the resulting serviceable number was too small to launch a campaign. Days were becoming weeks.

Meanwhile, there was internal pressure building to "just get something out there," since that is supposed to be one of the big advantages of email, right? I'm sure many of you can relate. So we conducted one campaign where the requisite pre-screening was NOT performed, because we wanted something out there, the universe was large, and the price was right. We also eventually found a vendor who could produce enough serviceable addresses for us to conduct a test campaign.

Actions and Results

Penn Media responded promptly to our request, and produced a proposal with about 200,000 email addresses of self-identified SOHO business people. Unfortunately, they could not accommodate the request for serviceability screening. In fact, they could not give us very many specifics on these addresses, outside of the fact that other DSL providers had used the SOHO category in the past. However, the price was right (only about 5 cents per record) so we were off - we needed to be out there and at that cost the endeavor was worth the risk.

The campaign outline was a text ad placed within editorial copy of various different lists, designed to bring a potential customer to a splash page where the discounted offer was reiterated and the visitor could fill in a follow-up form which was them emailed to DSL.net. As a result, we received 531 emails from potential customers who filled out the form and requested follow-up information, so the initial CTR was an under-whelming .26%.

Another result of note - approximately 15% of the unique visitors to the splash page took the second step of filling out the email form. Unfortunately, only about 20% of the respondents could be serviced, due to the lack of pre-screening. These 20% were converted to buyers at a percentage consistent with other marketing channels. We knew there would be large amounts of waste going in. But because the list was so inexpensive, and acquisition costs in the DSL market so high, the campaign more than paid for itself.

Another firm, Edirect, formerly PostMaster.com, eventually produced a list that could meet the requirements, and after pre-screening numbered approximately 43,000 - large enough for a statistically relevant test. (This company is not to be confused with PostMasterDirect.com, recently purchased by DoubleClick.)

With this campaign we were able to make some traditional selects, such as job title and size of business (over 10), and send the offer as a stand-alone text message, not as a text ad within editorial content. The cost was more than triple that of the Penn Media campaign - 16 cents per record. We split the universe, using the same offer but presenting it using two different creatives.

From this campaign we received 152 qualified leads, a seemingly lukewarm CTR of .35%, but all of these leads were serviceable, which was critical. The percentage of splash page visitors who filled out the email form was a very strong 22%. Interestingly enough, the two creatives pulled almost exactly equally.

Lessons/Observations

Based on these results, I would maintain that high email CTR's come from very clean house lists, used effectively. When you're dealing with B-to-B, rented lists, and with a relatively expensive product or service, successful response rates should be viewed as traditional direct mail - 1 to 2%. Even at these response rates, email brings a lot of advantages discussed earlier. From these efforts and others I have conducted, I would agree with the 60 - 25 - 15 split for effective email campaigns - 60% list, 25% offer, 15% creative.

It's vital to demand full transparency from vendors. Ask for selects and data cards as you would in a traditional direct mail campaign, and get answers why if these can't be provided. Clearly request and understand all elements of the campaign when working through brokers - rationale for recommendation, reporting, actual delivery of emails, schedule for delivery.

Don't waste time with vendors who don't listen to your needs. Don't forget to seed the list with yourself and others in the company. Don't totally depend on the vendor for tracking - it's easy to do the basics yourself. Negotiate hard if the CPM is high, and ask to test first. (Again, just like traditional D/M.)

Email marketing is still a very embryonic market. Being prepared lets you take full advantage of everything it has the potential to offer your client.

Christopher Parente designs and implements marketing and public relations programs for high tech companies in the Washington DC area, most recently DSL.net.